About the Role
This position involves owning the shared computer vision platform for a product that utilizes AI match analysis in youth football and is now venturing into a second industry. You will manage the end-to-end processes including camera ingestion, detection and tracking, OCR, classification, and the reporting pipeline. The goal is to raise accuracy on the sports models with real users while architecting deployments for the new industry leveraging the same foundational stack. Your responsibilities will encompass both the existing product, which is currently in production, and the upcoming deployment.
Ready to apply for roles like this?
Unlock the company name and direct application link. Subscribers get instant access to fresh jobs across Dubai, Abu Dhabi and Riyadh, many with visa support.
Unlock employer & apply directly
Responsibilities
- Manage the complete CV platform: camera ingestion, event detection, player tracking, OCR, classification, and reporting.
- Ship to a live product from day one: enhance accuracy on sports models with real user feedback.
- Structure the architecture: model selection, training strategy, annotation operations, evaluation methodology, and define the on-prem/cloud infrastructure.
- Oversee the first deployments in the new industry while integrating and hardening both products into a single cohesive platform.
- Provide technical leadership and build an engineering team under your supervision.
Hiring Process
- Introductory conversation focusing on a CV system you have built, discussing decisions, failures, and metrics.
- Technical deep-dive with senior engineers to assess your skills and experience.
- Case study involving real footage: specification and prototyping of a detection task.
- Final conversation with founders, followed by an offer.
- The entire process should take about two weeks for the right candidate.
What We Need to See
- 5+ years of experience in building production-grade computer vision systems, specifically deployed with real users.
- Proficiency in Python and PyTorch, with hands-on experience in modern detection and tracking techniques (e.g., YOLO family, ByteTrack/DeepSORT).
- Experience with video pipeline management, including RTSP/IP cameras, streaming inference, and fluency with ffmpeg.
- Proven track record of building datasets, running annotation operations, and objectively measuring model performance against ground truth.
- Experience fine-tuning open-source models (e.g., PaddleOCR/TrOCR) and replacing them with custom models trained on validated annotations, including managing an annotation-review loop.
- Familiarity with modern computer vision tools such as Roboflow or similar for dataset and annotation operations, as well as experiment tracking and model registries.
- An ownership mindset, with the ability to transform broad operational goals into functioning systems autonomously.
Strong Pluses
- Experience in OCR, action recognition, or pose estimation in a production environment.
- Background in sports, retail, industrial, or facility camera analytics.
- Knowledge of edge deployment (e.g., Jetson) and cost engineering for inference.
- Camera and IoT literacy: capability in sensor and lens selection, PoE networking, ONVIF/RTSP, NVRs, and weatherproof housing specifications.
Contract and Travel Details
- Full-time remote-friendly position, open to the best candidates regardless of location.
- Required working hours to overlap with Gulf time, including some necessary travel to Dubai for installations and milestone weeks.
- Competitive compensation package plus options.
